ProbFAST: Probabilistic Functional Analysis System Tool

Background  

The post-genomic era has brought new challenges regarding the understanding of the organization and function of the human genome. Many of these challenges are centered on the meaning of differential gene regulation under distinct biological conditions and can be performed by analyzing the Multiple Differential Expression (MDE) of genes associated with normal and abnormal biological processes. Currently MDE analyses are limited to usual methods of differential expression initially designed for paired analysis.     

BayGO: Bayesian analysis of ontology term enrichment in microarray data

Background  

The search for enriched (aka over-represented or enhanced) ontology terms in a list of genes obtained from microarray experiments is becoming a standard procedure for a system-level analysis. This procedure tries to summarize the information focussing on classification designs such as Gene Ontology, KEGG pathways, and so on, instead of focussing on individual genes. Although it is well known in statistics that association and significance are distinct concepts, only the former approach has been used to deal with the ontology term enrichment problem.     

Risk Factors Associated with Default from Multi- and Extensively Drug-Resistant Tuberculosis Treatment,Uzbekistan: A Retrospective Cohort Analysis

Background

The Médecins Sans Frontières project of Uzbekistan has provided multidrug-resistant tuberculosis treatment in the Karakalpakstan region since 2003. Rates of default from treatment have been high, despite psychosocial support, increasing particularly since programme scale-up in 2007. We aimed to determine factors associated with default in multi- and extensively drug-resistant tuberculosis patients who started treatment between 2003 and 2008 and thus had finished approximately 2 years of treatment by the end of 2010.

Methods

A retrospective cohort analysis of multi- and extensively drug-resistant tuberculosis patients enrolled in treatment between 2003 and 2008 compared baseline demographic characteristics and possible risk factors for default. Default was defined as missing ≥60 consecutive days of treatment (all drugs). Data were routinely collected during treatment and entered in a database. Potential risk factors for default were assessed in univariate analysis using chi-square test and in multivariate analysis with logistic regression.

Results

20% (142/710) of patients defaulted after a median of 6 months treatment (IQR 2.6–9.9). Factors associated with default included severity of resistance patterns (pre-extensively drug-resistant/extensively drug-resistant tuberculosis adjusted odds ratio 0.52, 95%CI: 0.31–0.86), previous default (2.38, 1.09–5.24) and age >45 years (1.77, 1.10–2.87). The default rate was 14% (42/294) for patients enrolled 2003–2006 and 24% (100/416) for 2007–2008 enrolments (p = 0.001).

Conclusions

Default from treatment was high and increased with programme scale-up. It is essential to ensure scale-up of treatment is accompanied with scale-up of staff and patient support. A successful first course of tuberculosis treatment is important; patients who had previously defaulted were at increased risk of default and death. The protective effect of severe resistance profiles suggests that understanding disease severity or fear may motivate against default. Targeted health education and support for at-risk patients after 5 months of treatment when many begin to feel better may decrease default.     

Multidrug-resistant tuberculosis treatment outcomes in Karakalpakstan, Uzbekistan: treatment complexity and XDR-TB among treatment failures

Background

A pilot programme to treat multidrug-resistant TB (MDR-TB) was implemented in Karakalpakstan, Uzbekistan in 2003. This region has particularly high levels of MDR-TB, with 13% and 40% among new and previously treated cases, respectively.

Methodology

This study describes the treatment process and outcomes for the first cohort of patients enrolled in the programme, between October 2003 and January 2005. Confirmed MDR-TB cases were treated with an individualised, second-line drug regimen based on drug susceptibility test results, while suspected MDR-TB cases were treated with a standardised regimen pending susceptibility results.

Principal Findings

Of 108 MDR-TB patients, 87 were started on treatment during the study period. Of these, 33 (38%) were infected with strains resistant to at least one second-line drug at baseline, but none had initial ofloxacin resistance. Treatment was successful for 54 (62%) patients, with 13 (15%) dying during treatment, 12 (14%) defaulting and 8 (8%) failing treatment. Poor clinical condition and baseline second-line resistance contributed to treatment failure or death. Treatment regimens were changed in 71 (82%) patients due to severe adverse events or drug resistance. Adverse events were most commonly attributed to cycloserine, ethionamide and p-aminosalicylic acid. Extensively drug resistant TB (XDR-TB) was found among 4 of the 6 patients who failed treatment and were still alive in November 2006.

Conclusions

While acceptable treatment success was achieved, the complexity of treatment and the development of XDR-TB among treatment failures are important issues to be addressed when considering scaling up MDR-TB treatment.     

ProbCD: enrichment analysis accounting for categorization uncertainty

Background  

As in many other areas of science, systems biology makes extensive use of statistical association and significance estimates in contingency tables, a type of categorical data analysis known in this field as enrichment (also over-representation or enhancement) analysis. In spite of efforts to create probabilistic annotations, especially in the Gene Ontology context, or to deal with uncertainty in high throughput-based datasets, current enrichment methods largely ignore this probabilistic information since they are mainly based on variants of the Fisher Exact Test.